summaryrefslogtreecommitdiffstats
path: root/sys/nfsserver/nfs_srvsock.c
blob: 71ee5f2aaff022db3d75cd1d88cee9be823309a9 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
/*
 * Copyright (c) 1989, 1991, 1993, 1995
 *	The Regents of the University of California.  All rights reserved.
 *
 * This code is derived from software contributed to Berkeley by
 * Rick Macklem at The University of Guelph.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)nfs_socket.c	8.5 (Berkeley) 3/30/95
 * $FreeBSD$
 */

#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");

/*
 * Socket operations for use by nfs
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/mount.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/protosw.h>
#include <sys/signalvar.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/vnode.h>

#include <netinet/in.h>
#include <netinet/tcp.h>

#include <nfs/rpcv2.h>
#include <nfs/nfsproto.h>
#include <nfsserver/nfs.h>
#include <nfs/xdr_subs.h>
#include <nfsserver/nfsm_subs.h>

#define	TRUE	1
#define	FALSE	0

static int nfs_realign_test;
static int nfs_realign_count;

SYSCTL_DECL(_vfs_nfsrv);

SYSCTL_INT(_vfs_nfsrv, OID_AUTO, realign_test, CTLFLAG_RW, &nfs_realign_test, 0, "");
SYSCTL_INT(_vfs_nfsrv, OID_AUTO, realign_count, CTLFLAG_RW, &nfs_realign_count, 0, "");


/*
 * There is a congestion window for outstanding rpcs maintained per mount
 * point. The cwnd size is adjusted in roughly the way that:
 * Van Jacobson, Congestion avoidance and Control, In "Proceedings of
 * SIGCOMM '88". ACM, August 1988.
 * describes for TCP. The cwnd size is chopped in half on a retransmit timeout
 * and incremented by 1/cwnd when each rpc reply is received and a full cwnd
 * of rpcs is in progress.
 * (The sent count and cwnd are scaled for integer arith.)
 * Variants of "slow start" were tried and were found to be too much of a
 * performance hit (ave. rtt 3 times larger),
 * I suspect due to the large rtt that nfs rpcs have.
 */
#define	NFS_CWNDSCALE	256
#define	NFS_MAXCWND	(NFS_CWNDSCALE * 32)
struct callout_handle	nfsrv_timer_handle;

static void	nfs_realign(struct mbuf **pm, int hsiz);	/* XXX SHARED */
static int	nfsrv_getstream(struct nfssvc_sock *, int);

int (*nfsrv3_procs[NFS_NPROCS])(struct nfsrv_descript *nd,
				struct nfssvc_sock *slp,
				struct thread *td,
				struct mbuf **mreqp) = {
	nfsrv_null,
	nfsrv_getattr,
	nfsrv_setattr,
	nfsrv_lookup,
	nfsrv3_access,
	nfsrv_readlink,
	nfsrv_read,
	nfsrv_write,
	nfsrv_create,
	nfsrv_mkdir,
	nfsrv_symlink,
	nfsrv_mknod,
	nfsrv_remove,
	nfsrv_rmdir,
	nfsrv_rename,
	nfsrv_link,
	nfsrv_readdir,
	nfsrv_readdirplus,
	nfsrv_statfs,
	nfsrv_fsinfo,
	nfsrv_pathconf,
	nfsrv_commit,
	nfsrv_noop
};


/*
 * Generate the rpc reply header
 * siz arg. is used to decide if adding a cluster is worthwhile
 */
struct mbuf *
nfs_rephead(int siz, struct nfsrv_descript *nd, int err,
    struct mbuf **mbp, caddr_t *bposp)
{
	u_int32_t *tl;
	struct mbuf *mreq;
	caddr_t bpos;
	struct mbuf *mb;

	nd->nd_repstat = err;
	if (err && (nd->nd_flag & ND_NFSV3) == 0)	/* XXX recheck */
		siz = 0;
	MGETHDR(mreq, M_TRYWAIT, MT_DATA);
	mb = mreq;
	/*
	 * If this is a big reply, use a cluster else
	 * try and leave leading space for the lower level headers.
	 */
	mreq->m_len = 6 * NFSX_UNSIGNED;
	siz += RPC_REPLYSIZ;
	if ((max_hdr + siz) >= MINCLSIZE) {
		MCLGET(mreq, M_TRYWAIT);
	} else
		mreq->m_data += min(max_hdr, M_TRAILINGSPACE(mreq));
	tl = mtod(mreq, u_int32_t *);
	bpos = ((caddr_t)tl) + mreq->m_len;
	*tl++ = txdr_unsigned(nd->nd_retxid);
	*tl++ = nfsrv_rpc_reply;
	if (err == ERPCMISMATCH || (err & NFSERR_AUTHERR)) {
		*tl++ = nfsrv_rpc_msgdenied;
		if (err & NFSERR_AUTHERR) {
			*tl++ = nfsrv_rpc_autherr;
			*tl = txdr_unsigned(err & ~NFSERR_AUTHERR);
			mreq->m_len -= NFSX_UNSIGNED;
			bpos -= NFSX_UNSIGNED;
		} else {
			*tl++ = nfsrv_rpc_mismatch;
			*tl++ = txdr_unsigned(RPC_VER2);
			*tl = txdr_unsigned(RPC_VER2);
		}
	} else {
		*tl++ = nfsrv_rpc_msgaccepted;
		/*
		 * Send a RPCAUTH_NULL verifier - no Kerberos.
		 */
		*tl++ = 0;
		*tl++ = 0;
		switch (err) {
		case EPROGUNAVAIL:
			*tl = txdr_unsigned(RPC_PROGUNAVAIL);
			break;
		case EPROGMISMATCH:
			*tl = txdr_unsigned(RPC_PROGMISMATCH);
			tl = nfsm_build(u_int32_t *, 2 * NFSX_UNSIGNED);
			*tl++ = txdr_unsigned(2);
			*tl = txdr_unsigned(3);
			break;
		case EPROCUNAVAIL:
			*tl = txdr_unsigned(RPC_PROCUNAVAIL);
			break;
		case EBADRPC:
			*tl = txdr_unsigned(RPC_GARBAGE);
			break;
		default:
			*tl = 0;
			if (err != NFSERR_RETVOID) {
				tl = nfsm_build(u_int32_t *, NFSX_UNSIGNED);
				if (err)
				    *tl = txdr_unsigned(nfsrv_errmap(nd, err));
				else
				    *tl = 0;
			}
			break;
		}
	}
	*mbp = mb;
	*bposp = bpos;
	if (err != 0 && err != NFSERR_RETVOID)
		nfsrvstats.srvrpc_errs++;
	return mreq;
}


/*
 *	nfs_realign:
 *
 *	Check for badly aligned mbuf data and realign by copying the unaligned
 *	portion of the data into a new mbuf chain and freeing the portions
 *	of the old chain that were replaced.
 *
 *	We cannot simply realign the data within the existing mbuf chain
 *	because the underlying buffers may contain other rpc commands and
 *	we cannot afford to overwrite them.
 *
 *	We would prefer to avoid this situation entirely.  The situation does
 *	not occur with NFS/UDP and is supposed to only occassionally occur
 *	with TCP.  Use vfs.nfs.realign_count and realign_test to check this.
 */
static void
nfs_realign(struct mbuf **pm, int hsiz)	/* XXX COMMON */
{
	struct mbuf *m;
	struct mbuf *n = NULL;
	int off = 0;

	++nfs_realign_test;
	while ((m = *pm) != NULL) {
		if ((m->m_len & 0x3) || (mtod(m, intptr_t) & 0x3)) {
			MGET(n, M_TRYWAIT, MT_DATA);
			if (m->m_len >= MINCLSIZE) {
				MCLGET(n, M_TRYWAIT);
			}
			n->m_len = 0;
			break;
		}
		pm = &m->m_next;
	}

	/*
	 * If n is non-NULL, loop on m copying data, then replace the
	 * portion of the chain that had to be realigned.
	 */
	if (n != NULL) {
		++nfs_realign_count;
		while (m) {
			m_copyback(n, off, m->m_len, mtod(m, caddr_t));
			off += m->m_len;
			m = m->m_next;
		}
		m_freem(*pm);
		*pm = n;
	}
}


/*
 * Parse an RPC request
 * - verify it
 * - fill in the cred struct.
 */
int
nfs_getreq(struct nfsrv_descript *nd, struct nfsd *nfsd, int has_header)
{
	int len, i;
	u_int32_t *tl;
	caddr_t dpos;
	u_int32_t nfsvers, auth_type;
	int error = 0;
	struct mbuf *mrep, *md;

	mrep = nd->nd_mrep;
	md = nd->nd_md;
	dpos = nd->nd_dpos;
	if (has_header) {
		tl = nfsm_dissect(u_int32_t *, 10 * NFSX_UNSIGNED);
		nd->nd_retxid = fxdr_unsigned(u_int32_t, *tl++);
		if (*tl++ != nfsrv_rpc_call) {
			m_freem(mrep);
			return (EBADRPC);
		}
	} else
		tl = nfsm_dissect(u_int32_t *, 8 * NFSX_UNSIGNED);
	nd->nd_repstat = 0;
	nd->nd_flag = 0;
	if (*tl++ != nfsrv_rpc_vers) {
		nd->nd_repstat = ERPCMISMATCH;
		nd->nd_procnum = NFSPROC_NOOP;
		return (0);
	}
	if (*tl != nfsrv_nfs_prog) {
		nd->nd_repstat = EPROGUNAVAIL;
		nd->nd_procnum = NFSPROC_NOOP;
		return (0);
	}
	tl++;
	nfsvers = fxdr_unsigned(u_int32_t, *tl++);
	if (nfsvers < NFS_VER2 || nfsvers > NFS_VER3) {
		nd->nd_repstat = EPROGMISMATCH;
		nd->nd_procnum = NFSPROC_NOOP;
		return (0);
	}
	nd->nd_procnum = fxdr_unsigned(u_int32_t, *tl++);
	if (nd->nd_procnum == NFSPROC_NULL)
		return (0);
	if (nfsvers == NFS_VER3) {
		nd->nd_flag = ND_NFSV3;
		if (nd->nd_procnum >= NFS_NPROCS) {
			nd->nd_repstat = EPROCUNAVAIL;
			nd->nd_procnum = NFSPROC_NOOP;
			return (0);
		}
	} else {
		if (nd->nd_procnum > NFSV2PROC_STATFS) {
			nd->nd_repstat = EPROCUNAVAIL;
			nd->nd_procnum = NFSPROC_NOOP;
			return (0);
		}
		/* Map the v2 procedure numbers into v3 ones */
		nd->nd_procnum = nfsrv_nfsv3_procid[nd->nd_procnum];
	}
	auth_type = *tl++;
	len = fxdr_unsigned(int, *tl++);
	if (len < 0 || len > RPCAUTH_MAXSIZ) {
		m_freem(mrep);
		return (EBADRPC);
	}

	/*
	 * Handle auth_unix;
	 */
	if (auth_type == nfsrv_rpc_auth_unix) {
		len = fxdr_unsigned(int, *++tl);
		if (len < 0 || len > NFS_MAXNAMLEN) {
			m_freem(mrep);
			return (EBADRPC);
		}
		nfsm_adv(nfsm_rndup(len));
		tl = nfsm_dissect(u_int32_t *, 3 * NFSX_UNSIGNED);
		/*
		 * XXX: This credential should be managed using crget(9)
		 * and related calls.  Right now, this tramples on any
		 * extensible data in the ucred, fails to initialize the
		 * mutex, and worse.  This must be fixed before FreeBSD
		 * 5.0-RELEASE.
		 */
		bzero((caddr_t)&nd->nd_cr, sizeof (struct ucred));
		nd->nd_cr.cr_ref = 1;
		nd->nd_cr.cr_uid = fxdr_unsigned(uid_t, *tl++);
		nd->nd_cr.cr_gid = fxdr_unsigned(gid_t, *tl++);
		len = fxdr_unsigned(int, *tl);
		if (len < 0 || len > RPCAUTH_UNIXGIDS) {
			m_freem(mrep);
			return (EBADRPC);
		}
		tl = nfsm_dissect(u_int32_t *, (len + 2) * NFSX_UNSIGNED);
		for (i = 1; i <= len; i++)
		    if (i < NGROUPS)
			nd->nd_cr.cr_groups[i] = fxdr_unsigned(gid_t, *tl++);
		    else
			tl++;
		nd->nd_cr.cr_ngroups = (len >= NGROUPS) ? NGROUPS : (len + 1);
		if (nd->nd_cr.cr_ngroups > 1)
		    nfsrvw_sort(nd->nd_cr.cr_groups, nd->nd_cr.cr_ngroups);
		len = fxdr_unsigned(int, *++tl);
		if (len < 0 || len > RPCAUTH_MAXSIZ) {
			m_freem(mrep);
			return (EBADRPC);
		}
		if (len > 0)
			nfsm_adv(nfsm_rndup(len));
	} else {
		nd->nd_repstat = (NFSERR_AUTHERR | AUTH_REJECTCRED);
		nd->nd_procnum = NFSPROC_NOOP;
		return (0);
	}

	nd->nd_md = md;
	nd->nd_dpos = dpos;
	return (0);
nfsmout:
	return (error);
}

/*
 * Socket upcall routine for the nfsd sockets.
 * The caddr_t arg is a pointer to the "struct nfssvc_sock".
 * Essentially do as much as possible non-blocking, else punt and it will
 * be called with M_TRYWAIT from an nfsd.
 */
void
nfsrv_rcv(struct socket *so, void *arg, int waitflag)
{
	struct nfssvc_sock *slp = (struct nfssvc_sock *)arg;
	struct mbuf *m;
	struct mbuf *mp;
	struct sockaddr *nam;
	struct uio auio;
	int flags, error;

	if ((slp->ns_flag & SLP_VALID) == 0)
		return;
#ifdef notdef
	/*
	 * Define this to test for nfsds handling this under heavy load.
	 */
	if (waitflag == M_DONTWAIT) {
		slp->ns_flag |= SLP_NEEDQ;
		goto dorecs;
	}
#endif
	auio.uio_td = NULL;
	if (so->so_type == SOCK_STREAM) {
		/*
		 * If there are already records on the queue, defer soreceive()
		 * to an nfsd so that there is feedback to the TCP layer that
		 * the nfs servers are heavily loaded.
		 */
		if (STAILQ_FIRST(&slp->ns_rec) && waitflag == M_DONTWAIT) {
			slp->ns_flag |= SLP_NEEDQ;
			goto dorecs;
		}

		/*
		 * Do soreceive().
		 */
		auio.uio_resid = 1000000000;
		flags = MSG_DONTWAIT;
		error = so->so_proto->pr_usrreqs->pru_soreceive
			(so, &nam, &auio, &mp, NULL, &flags);
		if (error || mp == NULL) {
			if (error == EWOULDBLOCK)
				slp->ns_flag |= SLP_NEEDQ;
			else
				slp->ns_flag |= SLP_DISCONN;
			goto dorecs;
		}
		m = mp;
		if (slp->ns_rawend) {
			slp->ns_rawend->m_next = m;
			slp->ns_cc += 1000000000 - auio.uio_resid;
		} else {
			slp->ns_raw = m;
			slp->ns_cc = 1000000000 - auio.uio_resid;
		}
		while (m->m_next)
			m = m->m_next;
		slp->ns_rawend = m;

		/*
		 * Now try and parse record(s) out of the raw stream data.
		 */
		error = nfsrv_getstream(slp, waitflag);
		if (error) {
			if (error == EPERM)
				slp->ns_flag |= SLP_DISCONN;
			else
				slp->ns_flag |= SLP_NEEDQ;
		}
	} else {
		do {
			auio.uio_resid = 1000000000;
			flags = MSG_DONTWAIT;
			error = so->so_proto->pr_usrreqs->pru_soreceive
				(so, &nam, &auio, &mp, NULL, &flags);
			if (mp) {
				struct nfsrv_rec *rec;
				rec = malloc(sizeof(struct nfsrv_rec),
			            M_NFSRVDESC, 
				    waitflag == M_DONTWAIT ? M_NOWAIT : M_WAITOK);
				if (!rec) {
					if (nam)
						FREE(nam, M_SONAME);
					m_freem(mp);
					continue;
				}
				nfs_realign(&mp, 10 * NFSX_UNSIGNED);
				rec->nr_address = nam;
				rec->nr_packet = mp;
				STAILQ_INSERT_TAIL(&slp->ns_rec, rec, nr_link);
			}
			if (error) {
				if ((so->so_proto->pr_flags & PR_CONNREQUIRED)
					&& error != EWOULDBLOCK) {
					slp->ns_flag |= SLP_DISCONN;
					goto dorecs;
				}
			}
		} while (mp);
	}

	/*
	 * Now try and process the request records, non-blocking.
	 */
dorecs:
	if (waitflag == M_DONTWAIT &&
		(STAILQ_FIRST(&slp->ns_rec)
		 || (slp->ns_flag & (SLP_NEEDQ | SLP_DISCONN))))
		nfsrv_wakenfsd(slp);
}

/*
 * Try and extract an RPC request from the mbuf data list received on a
 * stream socket. The "waitflag" argument indicates whether or not it
 * can sleep.
 */
static int
nfsrv_getstream(struct nfssvc_sock *slp, int waitflag)
{
	struct mbuf *m, **mpp;
	char *cp1, *cp2;
	int len;
	struct mbuf *om, *m2, *recm;
	u_int32_t recmark;

	if (slp->ns_flag & SLP_GETSTREAM)
		panic("nfs getstream");
	slp->ns_flag |= SLP_GETSTREAM;
	for (;;) {
	    if (slp->ns_reclen == 0) {
		if (slp->ns_cc < NFSX_UNSIGNED) {
			slp->ns_flag &= ~SLP_GETSTREAM;
			return (0);
		}
		m = slp->ns_raw;
		if (m->m_len >= NFSX_UNSIGNED) {
			bcopy(mtod(m, caddr_t), (caddr_t)&recmark, NFSX_UNSIGNED);
			m->m_data += NFSX_UNSIGNED;
			m->m_len -= NFSX_UNSIGNED;
		} else {
			cp1 = (caddr_t)&recmark;
			cp2 = mtod(m, caddr_t);
			while (cp1 < ((caddr_t)&recmark) + NFSX_UNSIGNED) {
				while (m->m_len == 0) {
					m = m->m_next;
					cp2 = mtod(m, caddr_t);
				}
				*cp1++ = *cp2++;
				m->m_data++;
				m->m_len--;
			}
		}
		slp->ns_cc -= NFSX_UNSIGNED;
		recmark = ntohl(recmark);
		slp->ns_reclen = recmark & ~0x80000000;
		if (recmark & 0x80000000)
			slp->ns_flag |= SLP_LASTFRAG;
		else
			slp->ns_flag &= ~SLP_LASTFRAG;
		if (slp->ns_reclen > NFS_MAXPACKET) {
			slp->ns_flag &= ~SLP_GETSTREAM;
			return (EPERM);
		}
	    }

	    /*
	     * Now get the record part.
	     *
	     * Note that slp->ns_reclen may be 0.  Linux sometimes
	     * generates 0-length RPCs.
	     */
	    recm = NULL;
	    if (slp->ns_cc == slp->ns_reclen) {
		recm = slp->ns_raw;
		slp->ns_raw = slp->ns_rawend = NULL;
		slp->ns_cc = slp->ns_reclen = 0;
	    } else if (slp->ns_cc > slp->ns_reclen) {
		len = 0;
		m = slp->ns_raw;
		om = NULL;

		while (len < slp->ns_reclen) {
			if ((len + m->m_len) > slp->ns_reclen) {
				m2 = m_copym(m, 0, slp->ns_reclen - len,
					waitflag);
				if (m2) {
					if (om) {
						om->m_next = m2;
						recm = slp->ns_raw;
					} else
						recm = m2;
					m->m_data += slp->ns_reclen - len;
					m->m_len -= slp->ns_reclen - len;
					len = slp->ns_reclen;
				} else {
					slp->ns_flag &= ~SLP_GETSTREAM;
					return (EWOULDBLOCK);
				}
			} else if ((len + m->m_len) == slp->ns_reclen) {
				om = m;
				len += m->m_len;
				m = m->m_next;
				recm = slp->ns_raw;
				om->m_next = NULL;
			} else {
				om = m;
				len += m->m_len;
				m = m->m_next;
			}
		}
		slp->ns_raw = m;
		slp->ns_cc -= len;
		slp->ns_reclen = 0;
	    } else {
		slp->ns_flag &= ~SLP_GETSTREAM;
		return (0);
	    }

	    /*
	     * Accumulate the fragments into a record.
	     */
	    mpp = &slp->ns_frag;
	    while (*mpp)
		mpp = &((*mpp)->m_next);
	    *mpp = recm;
	    if (slp->ns_flag & SLP_LASTFRAG) {
		struct nfsrv_rec *rec;
		rec = malloc(sizeof(struct nfsrv_rec), M_NFSRVDESC,
	            waitflag == M_DONTWAIT ? M_NOWAIT : M_WAITOK);
		if (!rec) {
		    m_freem(slp->ns_frag);
		} else {
		    nfs_realign(&slp->ns_frag, 10 * NFSX_UNSIGNED);
		    rec->nr_address = NULL;
		    rec->nr_packet = slp->ns_frag;
		    STAILQ_INSERT_TAIL(&slp->ns_rec, rec, nr_link);
		}
		slp->ns_frag = NULL;
	    }
	}
}

/*
 * Parse an RPC header.
 */
int
nfsrv_dorec(struct nfssvc_sock *slp, struct nfsd *nfsd,
    struct nfsrv_descript **ndp)
{
	struct nfsrv_rec *rec;
	struct mbuf *m;
	struct sockaddr *nam;
	struct nfsrv_descript *nd;
	int error;

	*ndp = NULL;
	if ((slp->ns_flag & SLP_VALID) == 0 || !STAILQ_FIRST(&slp->ns_rec))
		return (ENOBUFS);
	rec = STAILQ_FIRST(&slp->ns_rec);
	STAILQ_REMOVE_HEAD(&slp->ns_rec, nr_link);
	nam = rec->nr_address;
	m = rec->nr_packet;
	free(rec, M_NFSRVDESC);
	MALLOC(nd, struct nfsrv_descript *, sizeof (struct nfsrv_descript),
		M_NFSRVDESC, M_WAITOK);
	nd->nd_md = nd->nd_mrep = m;
	nd->nd_nam2 = nam;
	nd->nd_dpos = mtod(m, caddr_t);
	error = nfs_getreq(nd, nfsd, TRUE);
	if (error) {
		if (nam) {
			FREE(nam, M_SONAME);
		}
		free((caddr_t)nd, M_NFSRVDESC);
		return (error);
	}
	*ndp = nd;
	nfsd->nfsd_nd = nd;
	return (0);
}

/*
 * Search for a sleeping nfsd and wake it up.
 * SIDE EFFECT: If none found, set NFSD_CHECKSLP flag, so that one of the
 * running nfsds will go look for the work in the nfssvc_sock list.
 */
void
nfsrv_wakenfsd(struct nfssvc_sock *slp)
{
	struct nfsd *nd;

	if ((slp->ns_flag & SLP_VALID) == 0)
		return;
	TAILQ_FOREACH(nd, &nfsd_head, nfsd_chain) {
		if (nd->nfsd_flag & NFSD_WAITING) {
			nd->nfsd_flag &= ~NFSD_WAITING;
			if (nd->nfsd_slp)
				panic("nfsd wakeup");
			slp->ns_sref++;
			nd->nfsd_slp = slp;
			wakeup(nd);
			return;
		}
	}
	slp->ns_flag |= SLP_DOREC;
	nfsd_head_flag |= NFSD_CHECKSLP;
}

/*
 * This is the nfs send routine.
 * For the server side:
 * - return EINTR or ERESTART if interrupted by a signal
 * - return EPIPE if a connection is lost for connection based sockets (TCP...)
 * - do any cleanup required by recoverable socket errors (?)
 */
int
nfsrv_send(struct socket *so, struct sockaddr *nam, struct mbuf *top)
{
	struct sockaddr *sendnam;
	int error, soflags, flags;

	soflags = so->so_proto->pr_flags;
	if ((soflags & PR_CONNREQUIRED) || (so->so_state & SS_ISCONNECTED))
		sendnam = NULL;
	else
		sendnam = nam;
	if (so->so_type == SOCK_SEQPACKET)
		flags = MSG_EOR;
	else
		flags = 0;

	error = so->so_proto->pr_usrreqs->pru_sosend(so, sendnam, 0, top, 0,
						     flags, curthread/*XXX*/);
	if (error == ENOBUFS && so->so_type == SOCK_DGRAM)
		error = 0;

	if (error) {
		log(LOG_INFO, "nfsd send error %d\n", error);

		/*
		 * Handle any recoverable (soft) socket errors here. (?)
		 */
		if (error != EINTR && error != ERESTART &&
		    error != EWOULDBLOCK && error != EPIPE)
			error = 0;
	}
	return (error);
}

/*
 * NFS server timer routine.
 */
void
nfsrv_timer(void *arg)
{
	int s;
	struct nfssvc_sock *slp;
	u_quad_t cur_usec;

	s = splnet();
	/*
	 * Scan the write gathering queues for writes that need to be
	 * completed now.
	 */
	cur_usec = nfs_curusec();
	TAILQ_FOREACH(slp, &nfssvc_sockhead, ns_chain) {
		if (LIST_FIRST(&slp->ns_tq) &&
		    LIST_FIRST(&slp->ns_tq)->nd_time <= cur_usec)
			nfsrv_wakenfsd(slp);
	}
	splx(s);
	nfsrv_timer_handle = timeout(nfsrv_timer, NULL, nfsrv_ticks);
}
OpenPOWER on IntegriCloud